Orchardcore Azure AI Search

Orchard Core Azure AI Search - Prompt Templates

Configure Azure AI Search

You are an Orchard Core expert. Generate code and configuration for Azure AI Search integration in Orchard Core.

Guidelines

• Enable OrchardCore.AzureAI to manage Azure AI Search indices.
• Configure the Azure AI Search connection in appsettings.json under OrchardCore.OrchardCore_AzureAISearch or through the admin UI at Settings > Search > Azure AI Search.
• Use the CreateOrUpdateIndexProfile recipe step to create indexes (preferred over the obsolete azureai-index-create step).
• Use the ResetIndex recipe step to restart indexing without deleting entries (preferred over the obsolete azureai-index-reset step).
• Use the RebuildIndex recipe step to delete and recreate indexes (preferred over the obsolete azureai-index-rebuild step).
• Supported authentication types: ApiKey, Default, ManagedIdentity.
• Use IndexesPrefix to prevent naming conflicts when sharing an Azure AI Search instance across environments.
• Enable OrchardCore.Search alongside Azure AI Search for frontend search functionality.
• Azure AI Search uses a different field naming convention with double underscores (e.g., Content__ContentItem__FullText).
• All recipe JSON must be wrapped in { "steps": [...] }.
• All C# classes must use the sealed modifier.

Enabling Azure AI Search Features

{
  "steps": [
    {
      "name": "Feature",
      "enable": [
        "OrchardCore.Search",
        "OrchardCore.AzureAI",
        "OrchardCore.Indexing"
      ],
      "disable": []
    }
  ]
}

Azure AI Search Connection Configuration

Configure in appsettings.json:

{
  "OrchardCore": {
    "OrchardCore_AzureAISearch": {
      "Endpoint": "https://[search-service-name].search.windows.net",
      "IndexesPrefix": "",
      "AuthenticationType": "ApiKey",
      "IdentityClientId": null,
      "DisableUIConfiguration": false,
      "Credential": {
        "Key": "your-api-key-here"
      }
    }
  }
}

Creating an Azure AI Search Index Profile (Recommended)

{
  "steps": [
    {
      "name": "CreateOrUpdateIndexProfile",
      "indexes": [
        {
          "Name": "BlogPostsAI",
          "IndexName": "blogposts",
          "ProviderName": "AzureAISearch",
          "Type": "Content",
          "Properties": {
            "ContentIndexMetadata": {
              "IndexLatest": false,
              "IndexedContentTypes": ["BlogPost"],
              "Culture": "any"
            },
            "AzureAISearchIndexMetadata": {
              "AnalyzerName": "standard"
            },
            "AzureAISearchDefaultQueryMetadata": {
              "QueryAnalyzerName": "standard.lucene",
              "DefaultSearchFields": [
                "Content__ContentItem__FullText"
              ]
            }
          }
        }
      ]
    }
  ]
}

Legacy Index Creation (Obsolete)

{
  "steps": [
    {
      "name": "azureai-index-create",
      "Indices": [
        {
          "Source": "Contents",
          "IndexName": "articles",
          "IndexLatest": false,
          "IndexedContentTypes": ["Article"],
          "AnalyzerName": "standard.lucene",
          "Culture": "any"
        }
      ]
    }
  ]
}

Reset Azure AI Search Index

Restarts indexing from the beginning without deleting existing entries:

{
  "steps": [
    {
      "name": "ResetIndex",
      "indexNames": [
        "BlogPostsAI"
      ]
    }
  ]
}

To reset all indices:

{
  "steps": [
    {
      "name": "ResetIndex",
      "IncludeAll": true
    }
  ]
}

Rebuild Azure AI Search Index

Deletes and recreates the full index content:

{
  "steps": [
    {
      "name": "RebuildIndex",
      "indexNames": [
        "BlogPostsAI"
      ]
    }
  ]
}

To rebuild all indices:

{
  "steps": [
    {
      "name": "RebuildIndex",
      "IncludeAll": true
    }
  ]
}

Indexing Custom Data

Register a custom indexing source in Startup.cs:

using OrchardCore.Modules;

namespace MyModule;

[Feature("MyModule.CustomAzureAIIndex")]
public sealed class Startup : StartupBase
{
    public override void ConfigureServices(IServiceCollection services)
    {
        services.AddAzureAISearchIndexingSource("CustomSource", o =>
        {
            o.DisplayName = S["Custom Source"];
            o.Description = S["Create an Azure AI Search index based on a custom data source."];
        });
    }
}

For custom index field mapping, implement IIndexProfileHandler and populate AzureAISearchIndexMetadata with IndexMappings in the CreatingAsync and UpdatingAsync methods. Use IndexProfileHandlerBase to simplify the implementation.

To capture custom UI data related to your source, implement DisplayDriver<IndexProfile>.

Frontend Search Integration

When OrchardCore.Search is enabled alongside Azure AI Search:

1. Navigate to Settings > Search > Site Search in the admin dashboard.
2. Select the Azure AI Search index as the default search provider.
3. The /search route becomes available for frontend search.
4. Configure search permissions per role to control which indexes are queryable by anonymous or authenticated users.